Lung cancer is the leading cause of cancer mortality in the United States. Ethnic differences in lung cancer occurrence and survival are inadequately understood and may reflect both environmental and genetic influences. Elucidating the interplay of these factors in very high- and very low-risk populations will be crucial in developing novel intervention and prevention strategies. This study focuses on African-Americans who bear a disproportionate burden of lung cancer and Latinos who have very low rates. In this continuation of our ecogenetics study of lung cancer, we propose to expand recruitment of non-Latino Caucasian cases and controls to allow comparisons of smoking related cancer risk in three ethnic groups and test specific findings from the first study in a large population of African-Americans and Latinos in Northern California and will explore gene loci never before studied in minorities. We will collect blood or buccal specimens and interviews to achieve the following approximate sample sizes for the three groups:non-Latino whites (1200 cases, 1200 controls), African Americans (800 cases, 1600 controls), Latinos (450 cases, 900 controls). Cases will be identified through the Northern California Cancer Center (NCCC) Rapid Case Ascertainment program and through the entire Northern California Kaiser Permanente membership. All controls will be age-, race, county of residence and sex-frequency matched and recruited through Kaiser Permanente Northern California. We will establish a DNA bank and estimate the prevalence of genetic polymorphisms in carcinogen metabolism (e.g. CYPIA 1), DNA repair (e.g. XRCC1) and inflammation (MPO) as risk factors for lung cancer. Gene-environment interactions will be explored for smoking, diet and occupational risk factors. Population stratification and genetic admixture will be explored in the Latino and African American populations using both birthplace algorithms and population specific gene markers;these studies will help identify methodologic solutions to genetic studies in this diverse population. Previous associations of polymorphic traits with lung cancer among Caucasians are not readily generalizable to minorities without the further detailed study we propose. An especially unique contribution, our study represents the only broadly-based study of its kind among Latinos and includes a catchment area of 15 counties in Northern California. This will be only the second large genetic epidemiologic study of lung cancer in African-Americans;African-American men in California have an 11% lifetime risk of lung cancer.